Value-First AI seat
Operations
Mirror
Visual QA, interactive audit, and page review
Mirror is an AI seat of the Value-First Team — software with a defined role on the team, not a person.
About Mirror
Mirror is the visual QA eye for the Value-First Team. It audits the functional and visual integrity of web platforms — catching what breaks quietly, what renders wrong, and what confuses a first-time visitor before any of it reaches a real user.
Mirror
Visual QA & Interactive Audit Agent
“"Show me what the visitor sees -- and what happens when they click."”
Identity
Mirror is the Visual QA and Interactive Audit agent. While Lookout checks technical health, Mirror checks experience -- what the page looks like, how it responds to interaction, and whether the full user journey works end to end. Mirror uses Playwright to visit pages, capture what visitors see, and interact with elements the way a real user would.
Current State
An honest assessment of where this agent stands today.
What Works
- Playwright-based visual QA at multiple viewports
- Interactive audit with click-through testing for buttons, drawers, modals, navigation
- Accessibility basics: contrast, tap targets, focus indicators, keyboard navigation
- Before/after/reset screenshot protocol for every interaction
What Doesn’t Work Yet
- No baseline screenshot comparison for automated regression detection
- No scheduled execution -- requires manual trigger
Leadership Commentary
“Mirror is one of the most tangible agents we have -- it literally shows you what the visitor sees. The interactive audit capability is particularly valuable for complex components like the org chart, subway map, and portal walkthrough apps where click behavior is the experience.”
Delegation Contract
The observable, falsifiable standard this agent is held to.
Quality Bar
Every page review captures visual state and interaction behavior across viewports with before/after screenshots.
- Screenshots at desktop and mobile viewports
- All interactive elements tested with state changes
- Console errors captured
- Before/after/reset screenshots for interactions
- Accessibility basics checked
- No forbidden language
Invocation Triggers
Feedback Loop
Regression detection: issues that did not exist in previous reviews are regressions from recent changes. Before/after comparison across reviews is the regression signal.
Handoff
Lookout (complementary technical health checks)
Scope Boundary
Mirror checks visual and interactive quality. Lookout checks technical health (SEO, links). They complement each other.
The rest of the team
Other Value-First AI seats in Operations.
The team behind the work is on the record.
Every seat on the Value-First Team has a defined role, a public standard, and a name you can look up.